The study investigates how refined annotations versus model training changes affect pulmonary embolism segmentation performance. By re‑annotating 149 CT pulmonary angiography cases and evaluating two pretrained nnU-Net models, the authors find that improving annotation quality increases the Dice similarity coefficient (DSC) by 0.143–0.188, far exceeding the 0.028 DSC change from altering training datasets. A new human‑referenced benchmark model (nnPE) was trained and publicly released, though it performed below all annotators in paired comparisons.
arXiv:2608.24486v2 Announce Type: replace-cross Abstract: Reliable clot-volume quantification and subsequent risk assessment in pulmonary embolism depend on precise segmentation of emboli on computed...
The study evaluates how well different medical foundation models agree on image embeddings by measuring mutual k‑nearest‑neighbor overlap across 750,000 images from 14 datasets and 18 public models. Results show that label‑supervised models converge the least, sharing only 6–10% of nearest neighbors, whereas models trained with self‑distillation, masked image modeling, or contrastive pretraining share up to 80%. The authors suggest that developers can select pretraining objectives to control convergence and recommend testing unlabeled images for nearest‑neighbor consistency before replacing a model.
Med-AR introduces two autoregressive vision‑language models, Med‑AR‑8B and Med‑AR‑2B, pretrained on structured radiology reports, abnormality‑focused text, and region annotations to address long‑tailed chest X‑ray classification. The models outperform existing contrastive, self‑supervised, and supervised encoders—including Med‑CLIP, CheXFound, EVA‑Base, ARK, and BioViL‑T—across PadChest, MIMIC‑CXR, and CheXpert, achieving higher mean AUROC and AUPRC for head, medium, and tail findings and lower excess area under the risk‑coverage curve. Med‑AR also demonstrates improved selective‑prediction performance, with Med‑AR‑8B raising tail‑label mean AUPRC on MIMIC‑CXR from 0.1033 to 0.1441 and Med‑AR‑2B delivering the strongest discrimination on PadChest.
arXiv:2608.29348v1 Announce Type: new Abstract: Background: Patient details and acquisition metadata are important for clinical decisions, image quality control, and automated research pipelines, but...
The study investigates how image‑domain Poisson perturbations affect the classification of ischemic core/penumbra in non‑contrast CT slices. Using a CPAISD cohort, the authors compared direct ResNet‑18 classification with a denoising‑then‑classifying pipeline and found that denoising generally reduced performance. A prospective experiment with joint denoising‑classification models showed no statistically significant advantage over direct noisy classification, indicating limited robustness to Poisson noise.
arXiv:2609.37750v1 Announce Type: cross Abstract: Pulmonary embolism (PE) is a leading cause of cardiovascular mortality, yet the real-world performance of FDA-cleared AI detection models remains inc...
arXiv:2610.14226v1 Announce Type: cross Abstract: Purpose : To develop and evaluate CAPTOR, a deep-learning vision-language foundation model for oncologic chest-abdomen-pelvis (CAP) CT interpretation...
The study evaluates a deep‑learning model for pediatric pneumonia detection across chest X‑ray datasets from three countries, assessing discrimination, calibration, operating‑point transport, shortcut signals, and limited‑label recovery. Using a frozen DenseNet121 ensemble trained on Guangzhou data, the model achieved high internal AUROC (0.976) but performance dropped when applied zero‑shot to Bangladesh (AUROC 0.798) and Vietnam (AUROC 0.742). Limited‑label adaptation with Platt recalibration restored sensitivity but introduced significant specificity variability, highlighting the need to evaluate multiple performance dimensions in cross‑dataset transport studies.
arXiv:2607. 25589v1 Announce Type: cross Abstract: Medical-imaging AI benchmarks combine datasets, DICOM rendering, prompts, provider APIs, automated labels, statistical code, manuscripts, and repository releases.
arXiv:2609.38271v1 Announce Type: new Abstract: Morphological characteristics such as spiculation and lobulation play an important role in assessing pulmonary nodules on computed tomography (CT), par...
arXiv:2609.31461v1 Announce Type: new Abstract: Background: Large volumes of unlabeled knee MRI scans are available across repositories but remain insufficiently leveraged. We developed KneePreM, a k...
The study evaluates a deep‑learning model for pediatric pneumonia detection across three countries, testing not only discrimination but also probability calibration, fixed operating‑point transport, shortcut signals, and limited‑label recoverability. Using a DenseNet121 ensemble trained on Guangzhou data, the model achieved high internal AUROC (0.976) but performance dropped to 0.798 and 0.742 on Bangladeshi and Vietnamese datasets, respectively. Limited‑label adaptation with Platt recalibration restored sensitivity but introduced significant specificity variability, highlighting the need for comprehensive cross‑dataset evaluation.